Lawrence Berkeley National Laboratory Recent Work

Title A NEW AND CONVENIENT SYNTHESIS OF 2-DEOXY-D-RIBOSE FROM 2 ,4-O-ETHYLIDENE-D- ERYTHROSE

Permalink https://escholarship.org/uc/item/2sj58133

Author Hauske, J.R.

Publication Date 1979

eScholarship.org Powered by the California Digital Library University of California Submitted to Journal of Organic Chemistry LBL-8641 1". / , Prepri nt' "

A NEW AND CONVENIENT SYNTHESIS OF 2-DEOXY-D-RIBOSE FROM 2,4-0-ETHYLIDENE-D-ERYTHROSE.

James R. Hauske and Henry Rapoport

January 1979

Prepared for the U. S. Department of Energy under Contract W-7405-ENG-48

TWO-WEEK LOAN COpy lc",ECEIVED k.AWRENCE BERKSl.~:'{ LABORATORY This is a Library Circulating Copy ;,cr 281979 which may be borr()wed for two weeks. LIBRARY AND For a personal retention copy, call OOCUMENYS se:CYION Tech. Info. Division, Ext. 6782

,. \ \. -. DISCLAIMER

This document was prepared as an account of work sponsored by the United States Government. While this document is believed to contain correct information, neither the United States Government nor any agency thereof, nor the Regents of the University of California, nor any of their employees, makes any warranty, express or implied, or assumes any legal responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by its trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof, or the Regents of the University of California. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof or the Regents of the University of California. 2

A NEW AND CONVENIENT SYNTHESIS OF 2-DEOXY-D-RIBOSE

5 FROM 2,4-0-ETHYLIDENE-D-ERYTHROSE.

7 James R. Hauske and Henry Rapoport*

9 Department of Chemistry and Lawrence Berkeley Laboratory University of California, Berkeley, California 94720 1 0

1 1

1 2 Abstract: A new synthesis is described of 2-deoxy-D-erythro- 1 3 pentose [2-deoxy~D-ribose, 2-deoxy-D-arabinose (!)], starting 1 1+ from D-glucose. ,The synthesis proceeds through directolefination 1 5 of 2,4-Q-ethylidene-D-erythrose (~) by addition of the stabilized 1 6 ylides generated from dimethylphosphorylmethyl phenyl sulfide 1 7 (~) and the corresponding sulfoxide ~. These afford the key 1 e intermediates, thio-enol ether 7' and (l,S-unsaturated sulfoxide ~, 1 9 which when .subjected to mercuric ion-assisted hydrolysis gave 20 high yields of' 2-deoxy-D-ribose 0). This facil-e chain extension 2 1 ,of 2 required its existance as a monomer, and conditions effective 2 i for obtaining the mOnomer have been developed .. Detailed lH_ and 2 3 13 ' C-NMR studies of these compounds are presented. 21+

27 2

1 !1lt.~9.

2 The synthesis of 2-deoxy sugars is an active area of

3 investigation, since these compounds are frequent constituents of

q biologically important molecules. Specif~cally, our attention

5 was drawn to the synthesis of 2-deoxy-D-ribose (!), a necessary'

6 intermediate in nucleoside synthesis. Previously reported

7 syntheses of ! appeared unattractive due to relatively poor yields

8 (10-35%) as well as numerous and tedious purifications. The

9 strategies cortunon to essentially all these methods was either a I fn I 10 one-carbon degradation of D-glucose, or deoxygenation of an fn 2 2 1 1 appropriately substituted sugar derivative that.is less accessible

1 2 than glucose. An elegant example of the latter strategy is the 3 fn 3 1.3 ,recent synthesis ,. which proceeds via the deoxygenation of. the

1 q pentose arabinose. It occurred to us that a more convenierit

l.5:I?fo~edure might be evolved from a two-carbon degradation of

" ,~lucose ':;,16 with. subsequent chain elongation.

1 7 A useful intermediate readily applicable to this approach is fn 4 1 8 2,4-0-ethylidene-D-erythrose (2).4 However, since it has,been . . 5 6 fn 5;6 1 9 reported' that 2 exists primarily as a dimer 3, its usefulness for

20 chain elongation was in doubt. Thus, f~r 2 to 'serve as an efficient

2 1 intermediate for the synthesis of 1, two questions had to be answered: (1) is it possible to. prepare monomer 2; (2) would 2 be amenable to 2 2 nucleophilic addition and chain elongation via reaction with an ylide 2 3

2q in view of the free hydroxyl group, or would an attenuating blocking-deblocking routine be necessary? 25

2 6 Although we were unable to find' any reports which unequivocally detail the preparation of monomer 2, a mixture of monomer 2 and 27 - 3

1 dimer 3 has been reported to react with certain nucleophilic reagents

2 under carefully controlled conditions of solvent and temperature.

3 Unfortunately, olefination proceeded in only about 30% yield. The

.. limited nucleophilic additions were rationalized by presuming that

5 some concentration of monomer ~ existed, however, no spectroscopic 7 8 fn 7,8 6 evidence for monomer formation was presented.' Despite poor over-

7 all yield of olefin, these results are significant, since they

8 illustrate that 2,4-0-ethylidene-D-erythrose (~) is a potentially

9 suitable substrate for nucleophilic additions. Its potential i

10 rests on the ability to prepare essentially pure monomer and its

11 reaction with an appropriately substituted ylide, which upon

12 hydrolysis would directly afford!.. This report details the

13 successful implementation of this strategy.

1 ..

1 5 Results and Discussion

1 6 Preparation of 2,4-0-Ethylidene-D-erythrose and Its NMR

17 Evaluation. The synthesis of 2,4-0-ethylidene~D-erythrose (3)

18 proceeded via the very efficient periodate degradation of 4,6-0-

19 ethylidene-D-glucose 4.4 Rec~ystallization of the crude residue afforded crystalline material which exhibited the l3 and IH-NMR 2 0 c_ spectra shown in Figures I and 2, respectively. 2 1 No aldehydic

, 2 i resonance appears in either Figures 1 or 2, which is consistent with earlier reports of no aldehydic absorption either in its 2 3 , fn 9 l.n. f rare'd' spectrum 9 or l.n . 'ht e 1 H-NMR spectrum ofl.ts. acetates. 7 2 .. This lack of aldehydic resonance as well as the complexity of the 2 5 13 26 C-NMR spectrum, leaves no doubt that the product initially isolated is essentially 3. 2 7 4

Place structures 2, 3, 9 here

3 Presumably, there is an equilibrium between dimer ~ and ~ monomer 2,- and one might expect ylide addition. to shift the equili- 5 brium as monomer is consumed. However, addition of the ylides

6 generated from either phosphonium ,salt ~ or phosphonates ~ and 5

7 did not afford olefin, but rather resulted for the most part in the .. 8 recovery of starting material 3. Although the ethylidene erythrose

9 (~/~) has not previously served as a substrate for the synthesis of 5 10 2-deoxy-D-ribose (!), it has been reported ,6 to afford low yields

11 of ole fins upon treatment with certain phosphonate Ylides. In a

12 parallel fashion, we also obtained ole fins 7_ and 8_ in poor yield

13 upon treatment of the ethylidene erythrose with the phosphonate

1~ ylides 4y and 5y. In contrast, treatment of crystalline 3 with ethyl acetate 1 5 containing a catalytic amount of anhydrous afforded, after fn 10 1 6 aci~lO solvent removal, a solid residue which underwent olefin formation 1 7 in. excellent yield. Thus exposure of 3 to catalytic, anhydrous acid 1 8

resulted in transformation to the monomeric ethylidene erythrose, ~. 1 9 Removal of all traces of acid by continuous azeotropic distillation 20 with either benzene or toluene restored the original dimeric ·2 1 structure, 3. When this material was then resubmitted to the ylide 2 i reaction, it was e~sentially inactive. Although ylide would be 2 3 expected to destroy the residual acid and re-establishdimer 3, 2~ dimerization under these conditions is much slower than re~ction of 25 . the ylide with the monomer aldehyde 2 and high yields of olefins 26 result. 27 5

The ,13c_ and lH-NHR'spectra of monomeric 2,4-Q-ethylidene-D-

2 erythrose (~) are shown in Figures 3 and 4. The aldehydic resonance 13 . . 3 now is clearly evident in both spectra. The C-Nr-1R spectrum is

~ much simpler and indicates that less than 5% of dimer 3 is

5 present. However, it is complicated by an extra absorption, seven

6 carbon resonances appearing instead of the anticipated six. This

7 cannot be due to any diastereomers since such an explanation would

8 result in nonequivalence for more than one carbon resonance. We

9 attempted, therefore, to resolve this apparent anomaly by first

10 reducing 2 with sodium borohydride to the crystalline diol ~, 4 11 obtained in 96% yield. ,6 13 . 12 The C-NMR spectrum of diol 9 appears in Figure 5. Along with

13 the lack of an aldehydic resonance one sees the appearance of the

1~ newexocyclic hydroxymethyl absorption at 62.4ppm. More significant,

15 however, is the absence of the extra resonance which is present in

16 the spectrum of £ (Fig. 3)'. The six signals corresponding to the

1 7 six carbon nuclei were readily assigned as shown on the basis of

18 . chemical shift correlations as well as off-resonance decoupling

19 experiments. On the basis of this spectrum, diol 9 does not

20 evidence nonequivalence; that is, ~ is either diastereomerically

2 1 pure, or alternatively, the diastereomers have fortuitously 22 coincident chemical shifts. In order to test this latter possibility,

23 we attempted to induce nonequivalence in ~ by incremental addition

24 of a lanthanide shift reagent. When 9 was treated with Eu(fod)3 all

25 diol resonances shifted downfield in both the lH_ and 13C- NMR

2 6 spectra and there was no induced nonequivalence. Therefore, we

27 conclude that ~ is a single diastereomer which has an equatorial 6

6 1 ethylidene methyl. A similar conclusion was reached for 4-0-

2 acetyl-l,2-di-deoxy-3,5-0-ethylidene-l-nitro-D-erythro-pent-l-

3 enitol, prepared from 2. Since 9 is a single diastereomer, 2

~ must also be diastereomerically pure; therefore, the extra resonance

-5 appearing in Figure 3 must arise from another source. Although the

6 spurious resonance may arise from an impurity, this seems unlikely

7 in view of the Eu(fod)3 results; further a dynamic conformational

8 effect may be similarly ruled out. 13 9 Examination of the off resonance decoupled C-NMR spectrum of

1 0 3 in D2 0 exhibited doublets for all absorptions in the range of

1 1 80-100 ppm.· It would be reasonable to expect :the monomer-ic material

r 2 to form a hemi-acetal and this would account for the absorption

1 3 appearing at 89 ppm. A 13C- NMR spectrum of 2 in dried dioxane

1 ~ dld display seven absorptions; however the resonance at 89 ppm was

1 5 markedly diminished «15%) but still present, probably due to

1 6 residual moisture. This result provides support for the hemiacetal

1 7 hYpothesis.

1 8 Ylide Generation and Olefin Formation. Scheme I outlines

1 9 the synthesis of 2-deoxy-D-ribose (~). The olefination sequence

20 proceeded in moderate to high yield via either the phosphorane ~y

21 or the phosphonate ylides ~y and ~y. The success of the reaction

2 2 is critically dependent upon the relative basicity of the ylide

2 3 as well"as on the previously described method of preparation of

2~ 2,4-0-ethylidene-D-erythrose (2). ll fn 11 25 It has been demonstrated that stabili?ed ylides of type ~y and 5y add to a variety of hydroxy-substituted carbonyl substrates 26 '- 27 to afford either thioenol ehters or a,B-unsaturated sulfoxides. 7

1 Although stabilized ylides have been applied to the synthesis of

2 sugar moieties, they typically have been used to prepare phosphoryl- 5 12 fn 12 3 ated sugars. One exception, however, is the recent report of

~ the preparation of some deoxyhexoses via the stabilized ylide which

5 was generated from phenylthiomethyltriphenylphosphonium chloride (6).

6 When applied to 2, one equivalent of this ylide, 6y, afforded olefin

7 7. in a modest 27% yield; further, the addition of two equivalents

8 did not significantly improve the yield.

9 In an effort to improve this yield, we synthesized. the

10 stabilized phof?phonate ylides 4y and 5y. The addition o~ trimethyl

11 phosphite to a-chloromethyl phenyl sulfide afforded phosphonate ~,

12 which upon subsequent oxidation with either sodium metaperiodate or

13 peroxybenzoic acid was converted to dimethylphosphorylmethyl phenyl

1~ sulfoxide (5). Sequential treatment of either ~ or 2 with n-butyl-

0 0 15 lithium at -30 and -70 , respectively, and then with a dry tetra-

16 hydrofuran solution of 2,4-Q-ethylidene-D-erythrose (~), furnished

17 ole·fins ~ and ~, typically in 70-90% yields. Although the formation

18 of E and ~ isomers was possible for both 7 and ~, no effort was

19 made to separate them since any mixture of geometric isomers was

20 suitable for the next step of the synthesis. However, a qualitative

21 assessment of the isomeric (~~) composition was determined by fn 13, 1~:2 lH-NMR. Use was made of the established13 , 14 c·oupling constant for

23 the trans-vicinal protons in similar systems of 15-16 Hz, whereas,

2~ the cis-vicinal protons evidence a coupling constant of 10-11 Hz.

25 Further, there are well defined chemical shirt differences for.the 13 26E and Z isomers of similar thioe~ol ethers as well as a,S-unsatur- 14 27 ated sulfoxides. Typically, the .proton vicinal to the ci·te of 8

functionali ty appears at higher field for the E isomer o'f thioenol

2 ethers and at lower field for the E isomer of a,S-unsaturated

3 sulfoxides. We noted a similar trend, that is, the chemical shift

4 differences were on the order of 0.20 to 0.40 ppm between the ~ and

5 ! isomers and the respective coupling constants were approximately

6 15-16 Hz and 10-11 Hz. Utilizing these data as well as the tables fn 15 7 of chemical shifts foi olefinic substituents,15 we determined that

8 the ~ isomer was in excess by at least a 2/1 ratio.

9 , Although n-butyllithium proved to be a convenient base, we also

1 0 assayed a variety of other reagents and conditions for ylide 16 En 16,17 1 1 generation. For example, there have been several reports ,17,18 18 1 2 describing the utility of phase transfer catalysts for ylide

, 1 3 formation. These were applied in attempted ylide formation from 2.

1 4 With a large variety of phase transfer catalysts, ylide formation

1 5 did not proceed significantly with our substrates. Variations in solvent, temperature, and time as well as base all resulted in 1 6

1 7 essentially no reaction. Also, addition of a mixture of 18-crown-6, either catalytically or stoichiometrically, and potassium-t-butoxide 1 8 in tetrahydrofuran to a tetrahydrofuran solution of phosphonate 4 or 1 9 6 and the ethylidene erythrose 2, resulted in no detectable 2 0 - ' reaction. Finally, alkoxide proved unsatisfactory «10% yield) ; 2 1

2 i however, lithium diisopropyl amide at -70°, and sodium hydride or potassium hydride, both at -20°, all gave satisfactory results. All 2 3 of these procedures are inferior to ylide generation by addition of 24

2 5 n-butyllithium to either phosphonates 4 or 5 at -30° and -70°, respectively. 2 6

27 9

Conversion of Thio-substituted Olefins to 2-Deoxy-D-ribose (1).

2 All that remained to complete the synthesis of 2-deoxy-D-ribose (!)

3 was the hydrolyses of thioenol ether 7 and a,e-unsaturated sulfoxide

8 (Scheme II). We found that treatment of thioenol ether 7 with

5 aqueous sulfuric acid did not afford l.in satisfactory yields.

6 However, when the hydrolysis was conducted in.a stepwise fashion, 1

7 was obtained in excellent yields. Treatment of 7 first with

8 mercuric chloride in acetonitrile-water, or alternatively with a

9 mercuric oxide-mercuric chloride mixture in acetonitrile-water

1 0 afforded, after mild aqueous acid hydrolysis, 2-deoxy-D-ribose (1)

1 1 in greater than 75% yield. The a,e-unsaturated sulfoxide 8 was

1 2 converted to 1 by exposure to aqueous sulfuric acid; however, this

1 3 conversion did not proceed as readily as the thioenol ether pro-

1 4 cedure and afforded 1 in about 30% yield from 8.

1 5 Alternate Synthesis of 2-Deoxy-D-ribose. Since 2, containing . \ ~nto 1 6 a free hydroxyl group, had been successfully converted 1, we

1 7 considered the possibility of converting D-erytp,rose (11) into 1. The

1 8 utility of such an alternate approach stems from the added flexi

1 9 bility of performing the ylide addition in the absence of blocking groups.and would provide a model for olefination of underivatized 2 0 polyhydroxy aldehydes. Thus, hydrolysis of 2 with aqueous sulfuric 2 1

22 acid and subsequent treatment with the ylide 4y generated from

2 3 dimethylphosphorylmethyl phenyl sulfide (4) afforded the triol thio-

24 enol ether 12, which was not isolated but directly treated with a

2 s· mercuric oxide-mercuric chloride mixture in" acetonitrile-water

2 6 (Scheme III) •. The resulting crude product was recrystallized from

27 isopropanol to afford colorless, crystalline 1 in greater than 45% yield. 10

In 'summary-, we have described a convenient method for the

2 preparation of 2-deoxy-D-ribose (1) starting with D-glucose. Our

3 synthesis proceeds via monomeric 2,4-0-ethylidene-D-erythrose (2) , -- ,

4 as a key intermediate, and affords enhanced overall yields and

5 simplified isolation.

1 0

1 1

1 2

1 3

1 4

1 5

1 6

1 7

1 8

2 0

2,1

2 3

27 11 fn 19 ~~~~~~~~~~~~_~~~~~~~19

2 2',4-Q-Ethylidene-D-erythrose was prepared from D-glucose via , 4 3 periodate oxidation of 4,6-Q-etq.ylidene-D-glucose. Its properties 4 ~ were in agreement with those reported ,5,6 and established that the

5 cr},stallized material was primarily dimer 3 with less than 5% of

6 IHNMR(D 0) 01.37 (d, CCH ), 3.35-3.95 (m, OCH), 4.25 monomer~. 2 3 13 ' ' 7 (td, OGH ), 4.75-5.50 (m', HO and 02CH); 'C NMR (dioxane) 0 20.13 2

8 (CH ),: 61.19 (CHOH, d in the off-resonance spectrum), 67.04; 67.88, 3 9 68.32 (CH 0, t in the off-resonance spectrum), 70.29, 71.50, 75.89, 2

1 0 78. 44, 80. 59, 90. 62, 91. 68 (CH 0), 95. 85, 97. 42, 100. 23, 101. 2 -1 0 "4' 0 1 1 [H~(CH3) °2 ]; IR 3435 (OH) cm ,; mp 150-151 (lit. mp 149-150 ); 4 0 1 2 [cd~5 -36.20 (equilibrium, £. 8.2, H20) [lit. [o.]~O -36.8 (equilibrium, £,8.25, H 0)]. 1 3 2 The dimer ~, prepared above,was dissolved in ethyl acetate

1 5 (20 mL/g) and several drops of glacial acetic acid or 100% phosphoric lO 16 acid were added. The resulting solution was heated at 90 0 for

17 20 min, after which it was quickly cooled to 25 0 and the ethyl

18 acetate was evaporated at this temperature. The solid residue thus

19 obtained was used without further treatment in the olefination

20 reactions. Its properties indic~ted that it'was primarily the ,1 21 monomeric 2,4-Q-ethylidene-D-erythrose (~): H NMR (D 0) 0 1.25 2 (d, CCH ), 3.47-3.83 (m,OCH), 4.17 (dd, OCH ), 4.86-5.28 (m, HO 2 i 3 2 13 ' and 02CH), 8.47 (s, CHO); C NMR (dioxane) 0 20.32 (CH ), 62.09 2 3 3 (CHOH, d in off-resonance spectrum), 70.21 (CH 0, t in'off-resonance 2

25 spectrum), 82.65 (OCH,d in off-resonance spe~trum),' 89.0, 99.88

26 (02CH , d in off-resonance spectrum), 171.45 (CHO" d in off-resonance 2., spectrum); IR 3433 (OH), 1718 (C=O) cm-l . 12

a-Chloromethyl Phenyl Sulfide. ,A solution of sulfuryl

2 chloride (0.21 mol, 28.4 g) in CH 2C1 2 was slo~ly added over a 2 h

3 period to a heated (reflux) solution of methyl phenyl sulfide

.. (0.20 mol, 25 g) in CH C1 . After addition was complete, the 2 2

5 mixture was heated at reflux for another,hour and allowed to stand

6 at room temperature overnight. The major portion of the solvent , 7 was evaporated and the residue distilled at reduced pressure: yield, 20 fn 20 8 92%; bp 110-115°C (20 mm) [lit. bp 104°C (12 mm)]; NM~ (CDC1 3)

9 <5 4.75 (s, ClCH2 ), ?25 ppm (m, C6H5 ).

1 0 Dimethylphosphorylmethyl Phenyl Sulfide (4). A mixture of

11 a-chloromethyl phenyl sulfide (41 g, 0.26 mol) and trimethyl

12 phosphite (4 mL/g of sulfide) was heated at ISO-160°C for 36 h. The

1 3 reaction mixture was allowed to cool to room temperature and the

1 .. unreacted trimethyl phosphite was evaporated (40-50 mm) with heating.

1 5 The residue was then fractionally distilled through a 7 in. vigreu'x

1 6 column at high vacuum to afford a colorless liquid: yield, 83%;

17 bp l40-l43°C (0.5 rom); NMR (CDC1 ) <5 3.5 [d, CH P(O)], 3.71 [d, 3 2

1 8 P H (CH 30) 2 ], 7.29 ppm (mul ~iplet, C6 5).·

19 DimethylEhosphorylmethyl Phenyl Sulfoxide (5). To a solution

20 of dimethylphosphorylmethyl phenyl sulfide (4, 23.2 g, 0.10 mol)

21 in acetone (130 mL) and water (70 mL), cooled to -5°C, was added

22 dropwise a solution of sodium metaperiodate (22.5 g, 0.11 mol, in

23 enough water to dissolve the periodate) over 1 h at -5 to O°C. The

2 .. solution was stirred at O°C for 4 hr and allowed to stand at 5°C

25 for 24 h, the' precipitated sodium iodate was 'removed, and after

26 evaporation of the acetone, the solution was extracted with CHCI • 3

27 Drying over magnesium sulfate and, evaporating the CHC1 gave sulfoxide 3 13

1 ~ as a liquid, which was purified by column chromatography on silica,

2 eluting .with CHC1 /CH 0H, 9/1: yield, 95%; NMR (CDC1 ) 15 3.29 3 3 3 3 and 3.5 [AB part of an ABX system, CH 2P (0)] 3,.72 [overlapping " doublets, (CH 0 P]' 7.51 ppm (multiplet, C H ). 3 2 6 5 fn 21 5 Phenylthiomethyltriphenylphosphonium Chloride (~) .21 To

.6 triphenylphosphine (200 g, 0.76 mol) dissolved in benzene (400 mL)

7 was added dropwise a-chloromethyl phenyl sulfide (122 g, 0.78 mol)

B with stirring at room temperature. After addition was complete,

the mixture was heated at reflux with stirring for 10 days. Cooling 9 to room te.mperature and standing gave the crystalline phosphonium 1 0 salt 6 in 43% yield: mp 154-156°. 1 1 Thioenol Ether 7 via Phosphonate 4.· To a tetrahydrofuran 1 2

(12~5 mL) solution of dimethylphosphorylmethyl phenyl sulfide (4_, 1 3 2.32 g, 0.01 mol)" cooled to -70°C, was added over 2 h a -hexane 1 " solution of n-butyllithium ( 0 • Oil mo 1) . The mixture was stirred 1 5

for 1 hr at -70°C and additionally for 2 h at -40°C'. then a tetra- 1 6 , hydrofuran (16 mL) solution of 2,4-Q-ethylidene-D-erythrose (2, 1 7 2g, 0.01 mol), prepared as described, was addeddropwise to the 1 B mixture maintairied at -40°C. After addition was complete, the 1 9 contents were stirred at -40°C for 2 h, then 'allowed to slowly, warm 2 0 to room temperature and stirred at room temperature overnight... The 2 1 residue after evaporation of the solvents was treated with water 22 23 (25 mL) and extracted with CHCl (5xl5 mL) and the CHCl was washed 3 3 with water (IS mL), dried over magnesium sulfate and evaporated 2 " to afford thioenol ether 7 in 70-7S% yield: .NMR (CDC1 )0 1.37 (bd, 25 3

26 CHC!!.3)' 3.S2-4.31 [OC!!.C!!.{0!!.)CH 0], 4.75 (bqt, CH C!!.), 6.15 (d ofd, 2 3

27 C6HSSCHC!!.), 6.72 (d, C6H5SC!!.CH), 7.35 (m, C6HS); HRMS, calcd. for 14

+ 1 . C13H1603S .(M ), 252.0815, found 252.0798 ..

2 a,B-Unsaturated Sulfoxide~. To a dry tetrahydrofuran (20 mL)

3 solution of dimethylphosphorylmethyl phenyl sulfoxide '(5, 2.48 g,

~ 0.01 mol), cooled to -70°C, was slowly added over 3 h a hexane

" s solution of n-butyllithium (0.011 mol). The mixture was stirred

6 for 4 h at -70°C, after which a dry tetrahydrofuran (25 mL) solution

7 of 2,4-Q-ethylidene-D-erythrose (2, 2 g, 0.01 mol) was added.

8 After addition was complete, the contents were stirred for 2 h at

9 -70°C, then allowed to slowly warm to room temperature and stirred . , 10 at room temperature overnight. The .reaction mixture was treated

11 exactly as in the isolation of thipenol ether 7 and resulted in a

12 75-87%, yield of a, sulfoxide NMR (CDCl ) 0 1.35 B~unsaturated ~: 3

13 (d of d, CHC!!.3)' 3.40-4.25 [m,oC!:!,C!!,(OH)CH20], 4.8,0_ (b q, CH 3C!!.),

1~ 6.55 [d, S(O)CH]" 7.00 [d of d, S(O)CHC!:!), 7.55 (m,C6H5 ); HRMS, ...... ' .' +. 15 calcd. for Cl3H1604S (M ), 2~6.0866, found 236.0861.

1 6 Thioenol Ether 7 via Phosphonium Salt 6. To a solution of

17 phenylthiomethy'ltriphenyl phosphonium chloride (6',4.20 g, 0.01 mol)

18 in tetrahydrofuran (75 mL), cooled to -25°C, ~a~ added dr?pwise over

19 3 ha hexane solution of n-butyllithium (0.011 mol). Stirring was

20 continued for 2 h with the temperature maintained at -25°C, after

21 which a tetrahydrofuran (30 mL) solution ot: 2,4-Q-ethylidene-D-

22 erythrose (2, 2 g, 0.01 mol) was slowly added over 2 h. The con-

23 tents were stirred an additional 2 h at 25°C, allowed to warm to

24 room 'temperature and stirred at room temperature overnight. The

25 reaction mixture was treated exactly as above and afforded a 27%

26 yield of thioenol ether Z, identical with material prepared from

27 phosphonate 4. 15

2,4-Ethylidene-0-erythritol (~) was prepared in 96% yield from 4 2 } in a similar fashion to that reported ,6 for the preparation of 9

3 from 2: mp 100° (lit. mp 99.S-100.so,498.5-100!506); [a.]~2 -54.3 . 24· 4· 25 6 4 (c 2.0, H 0) [l~t. [0.]0-54.7 (c 2.0, H 0) , [0.]0 -51.1 (c 0.67, H 0) ]. 2 2 2 5 2-0eoxy-0-erythro-pentose (2-0eoxy-D-rib?se, ]J. Procedure A.

6 To a solution of thioenol ether (I, 1 g, 3.96 romol) in acetonitrile/

7 water (3/1, 35 mL) was added a solution of mercuric chloride

8 (2.14 g, 7.92 romol) in acetonitrile/water (4/1, 35 mL) and the

9 resulting cloudy mixture was stirred at 50°C for 30 h. The mixture

10 was filtered through Celite~ the insoluble portion was washed

11 thoroughly with warm pyridine, and the filtrate was evaporated to

12 a crude residue which was treated with trifluoroacetic acid (O.OlM,

13 10 mL) at 6°C for 36 h. The remaining acid was removed by repeated

14 azeotropic distillation with water at reduced pressure, and the

15 residue, upon recrystallization from isopropanol, afforded colorless, 22 f'n 22 16 crystalline ~ in 92% yield: mp 92-94° (lit. mp 95-98°,1 96_98° ); -57.2 ° (c 0.5, H 0) [lit. -57.6° (c 1.1, H 0)1, -58° 1 7 [a.]~5 2 [a.]~O 2 fn 23 (c 1.6, H 0)22]; IH and 13CNMR identical to those reported. 3 ,23 1 8 2

19 Procedure B. To a solution of thioenol ether (I, 1 g, 3.96

20 romo!) in acetonitrile/water (4/1, 50 mL) was ,added with stirring

21 a mixture of yellow mercuric oxide (1.07 g, 3.46 romol) and mercuric

22 chloride (0.86 g, 3.96 romol).The contents were heated to 35°C for

23 16 hr, cooled to room temperature, and the product was isolated

24 exactly as in procedure A. The yield of pure 2-deoxy-0-ribose (1)

25 was 88%.

26 Procedure C. via O-Erythrose (II). To a vigorously stirred

27 solution of O-erythrose (11, 120 mg, 1 romol, prepared as described4 ) 16

1 in DMF (IS mL) or in DMSO (15 mL) at ~40° was added dropwise a

2 THF solution, also at -40,0" of the ylide iY [prepared by the

3 addition of n-butyllithium (2 mmol, in hexane) to a THF solution

,,(7 mL) of dimethylphosphorylmethyl phenyl sulfide (i, 0.46 g, 2 mmol)].

0 5 Af'ter addition was completed, the mixture was stirred for 2 h at -40 ,

6 allowed to warm to 20°, and stirred at 20° overnight. The mixture

7 was concentrated and treated with mercuric chloride, and the prcx:1uct was

8 isolated 'as described in Procedure A, affording crystalline 1 in

9 45% yield.

1 0

11 Acknowledgement. This research was supported in part by

12 the Division of Biomedical and Environmental Research of the

13 U.S. Department of Energy, Contract W-7405-ENG-4B.

1 "

1 5

1 6 '

1 7

1 8

1 9

2 0

2 1

2 3

27 17

References and Notes ------~------2 (l) H. W. Diehl ,and H. G. Fletcher, Jr., Biochem. Prepns,

3 9, 49 (1961). "(2) (a) J. Tomasz, Acta Chimica (Budapest), 79, 263 (1971);

5 (b) R. M. de Lederkremer and L.F. Sola, Carbohydrate Res.,

6 40, 385 (1975).

7 (3) M. Y. H. Wong and G. R. Gray, J. Am. Chern. Soc., 100, 3548

e (1978) . 9 (4) R. Barker and D. L. MacDonald, J. Am. Chern. Soc., --82, 2301 1 0 (1960) • 1 1 (5) H. Paulsen, W. Bartsch and J. Thiem, Chern. Ber., 104,--- 2545 1 2 (1971) .

1 3 (6) K. D. Carlson, C. R.Smith, Jr. and 1. A. Wolff, Carbohydrate

1 " Res., 13, 391, 403 (1970).

1 5 (7) Although references 6 and 8 have IH-NMR data for the acetates\

1 6 of dimer J, no IH-NMR or 13C- NMR data for the parent compound

17 'were reported.

, 1 e (8 ) R. Andersson, O. Theander and ,E. Westerlund, Carbohydrate',,'

1 9 Re s ., 61, 501 ( 19 7 8) . 2 0 (9) R. Schaffer, J. Am. Chern. Soc., 81,-- 2838 (1959). 2 1 (10 ) Initially anhydrous gaseous hydrochloric acid was used;

2 i s~bsequently we ,found it much more convenient to use either

2 3 100% phosphoric acid or glacial acetic acid.

(11) W. S. Wadsworth, Jr., Org. Reactions, 25, 73 (1977). 2 "

2 5 (12) H. J. Bestmann and J. Angerer, Tetrahedron Lett., 3665 (1969).

2 6 (13) E. J. Cory and J. 1. Shulman, J. Org. Chern., 35, 777 (1970)., 18

1 (14) M. Mikolajczyk, S. Grzejszczak and A. Zatorski, J. Org.

.., 2 Chern., 1Q, 1979 (1975).

3 (15) (a) C. Pascual, J. 1-1eier and N. Simon, Hel v. Chima Acta.,

.. 1~, 164 (1966); (b) U. E. Matter, C. Pascual, E. Pretsch,

!i A. Pross, W. Simon and S. Sternhell, Tetrahedron, ~~, 691

6 (1969).

7 (16) M. Mikolajczyk, S. Grzejszczak, W. Midura and A. Zatorski,

B Synthesis, 278 (1975).

9 (17) E. D'Incan and J. Seyden-Penne, Synthesis, 516 (1915).

1 0 (18) E. V. Dehmlow, Angew Chem., Int'l. Ed. Engl., 16, 493 (1977).

1 1 (19) All reactions were performed under a nitrogen atmosphere.

1 2 Solvent evaporations were carried but in vacuo using a

1 3 Berkeley Rotary Evaporator. All melting points are uncor-

1 .. rected. Proton NMR? spectra were recorded on a Varian EM-390,

1 5 HR-220 or Bruker WH-180 .. Carbon NMR spectra were recorded on

1 6 a Bruker WH-180; IR spectra were recorded on a Perkin-Elmer

1 7 337 Spectrophotometer as mulls. CEC-l03 and 110B spectrometers

1 e were used for determining mass spectra.

1 9 (20 ) F. Bordwell and B. Pitt, J. Am. Chern. Soc., 77, 572 (1955).

20 (21) Compound~ has been previously reported (ref. 12), however

2 1 neither experimental details nor characterizing data were

22 reported.

23 (22) R. Deriaz, W. Overend, M. Stacey, E. Teece and L. Wiggins,

2 .. J. Chern. Soc., 1879 (1949).

25 (23) E. Breitmaier, G. Jung and W. Voelter, Chimia, 26,136 (1972).

27 19

I. ! o

I C\I

oI I ~ I o 20

Scheme. I. Reaction of. 2,4-0-Etbylidine-D-erythrose (~) with

'. Various Ylides~y, ~y and ~y. 11- . (CH30)2PCHSCsH5° 4Yt 70% o 0 11- II (CH30)2PCHSC6H5\.. ~O~ /OH 0 5y,87. /0.. .. O~ ~~X

H (CSH5)3P=CHSCSH5 7, X = SCsH5 2 6Yt 27% 8· X = SC6H5 t. II . o

N ..oJ 22

Scheme· II. ConversIon of Thioenol· Ether 7 to 2-Deoxy-D-ribose (1). ~ - 23

I o

.0 . . -

I - o

:r: o 24

Scheme III. Conversion of 2,4-0-Ethy1idene-D-erythrose (~) to 2-Deoxy-O-ribose (1) via D-Erythrose. (11)..... • CHO . I H+ H-C-OH ~~H 'i: ~ . I H-C-OH H I 2 CH2OH. II

11- (CH30)2PCHSCsH5° ..~ .~ 4y

CHO :"' . I H, ~-"YSCsH5' CH C I . 2 Hg++' I H-C-OH H-C-OH l,~ I ~ . I H-C~OH H-C-OH I I CH20H ~H20H I 12

-;"'. '~.

N (J1 26

The 13c NMR'spectrum of 2,4 ... 0-ethylidene-:-O-.erythrose dimer (3).,. •

, . '" .... " .

" .... . , ,- .-

, .

,;, ..

......

r '- ~ 27

...",," I- 0 0 a ! (\J I 0 ~ ~..w.L Q) c.;r a)(r- . .2 to "tl

E 0- 00. o

o (\J

o ro 28

Figure 2. The 18 NMR spectrum· of 2,4-0-ethy.lidene-0-erythrose ...... ' - . di~er . (~) in O ° (external TMS). .2 . ., 29

_'. -1.

",; .... " .

o------~~======~~

E 0. a. .0 <.D :r: 0

o 30

Figure 3·. The 13C NMR spectrum of 2.,4-0-ethylidene-D-erythrose ...... """' ...... (2).. resulting fromacici-ethyl acetate treatment • dioxane 67.4 4

'. 6 2 20.3 V OH . 5 2 '1' I'OH 82.6 4, " 99.8 H l' 89.011 70.21 1 162.0

1 I 171.4 .

~~~~~~~~~ ~~~" , .:.

I I .1 1 I I I I I. I I I I I I II I I I n r l'-T-II~-T-r r ---r 1 I· I W 180 . 160 140 120 100 80 60 40 20 0 I-' ppm 32

~!2~~=_~. The IH NMR spectrum at 180 !1Hz in .D20 of. 2,4-0- ethylidene-D-erythrose (2) resulting from acid~ ethyl acetate treatment. 33

en ~ t- O lO (\J.. ~.. 0

·0 0 LO

N :c o:::C 0 0 0 ..Q :c

r- ~.

Figure 5. The llC NMR spectrum of.2,4-0-ethylidene-D- ... -...... - .... erythritol (~).

\ . 35 en :lE I- 0

~ d C\J 0 (Q C\.I

o ¢

E a. a.

o C\.I

o <.0

o 00 This report was done with support from the Department of Energy. Any conclusions or opinions expressed in this report represent solely those of the author(s} and not necessarily those of The Regents of the University of California, the Lawrence Berkeley Laboratory or the Department of Energy. Reference to a company or product name does not imply approval or recommendation of the product by the University of California or the U.S. Department of Energy to the exclusion of others that may be suitable. ,~ " TECHNICAL INFORMATION DEPARTMENT LAWRENCE BERKELEY LAB ORA TOR Y UNIVERSITY OF CALIFORNIA BERKELEY, CALIFORNIA 94720